Marios Richards' Data Science Blog

Mark Pack's Pollbase Analysis

In [1]:

# %matplotlib inline
import numpy as np, pandas as pd, matplotlib.pyplot as plt, seaborn as sns
import pickle, os, gc, re

from utility import *
BES_data_folder = '../BES_analysis_data/'
Pollbase_folder = create_subdir(BES_data_folder,"Pollbase")

In [13]:

from bokeh.models import ColumnDataSource
from bokeh.plotting import figure, show,  output_notebook
from bokeh.layouts import column, row
from bokeh.embed import components

In [3]:

output_notebook()

Loading BokehJS ...

In [14]:

components()

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-14-dd54e8e858b8> in <module>
----> 1 components()

TypeError: components() missing 1 required positional argument: 'models'

In [4]:

df = pd.read_csv(Pollbase_folder+"MarkPackPollBase_flat_v5.csv",
                 parse_dates=['Fieldwork_Start_Date', 'Fieldwork_Finish_Date',
       'Fieldwork_Midpoint_Date','Fieldwork_Start_check_date'],index_col="Unnamed: 0")

In [5]:

df.dtypes

Out[5]:

Year                                   int64
Month                                 object
Fieldwork_Start_Day                  float64
Fieldwork_Finish_Day                 float64
Publication Day                      float64
                                   ...      
sheet_row_number                     float64
Fieldwork_Start_check_date    datetime64[ns]
Fieldwork_Start_Date          datetime64[ns]
Fieldwork_Finish_Date         datetime64[ns]
Fieldwork_Midpoint_Date       datetime64[ns]
Length: 82, dtype: object

In [ ]:

In [12]:

source = ColumnDataSource(df)

p1 = figure(x_axis_type="datetime",x_axis_label = "Fieldwork_Midpoint_Date",y_axis_label="Con",
            plot_width=600, plot_height=250, title="Conservative Voteshare",tools='box_select,lasso_select,pan,wheel_zoom,box_zoom,reset')
p1.scatter('Fieldwork_Midpoint_Date', 'Con', source=source, color='blue')
p2 = figure(x_axis_type="datetime",x_axis_label = "Fieldwork_Midpoint_Date",y_axis_label="Lab",
            plot_width=600, plot_height=250, title="Labour Voteshare",tools='box_select,lasso_select,pan,wheel_zoom,box_zoom,reset')
p2.scatter('Fieldwork_Midpoint_Date', 'Lab', source=source, color='red')
p3 = figure(x_axis_type="datetime",x_axis_label = "Fieldwork_Midpoint_Date",y_axis_label="Lab",
            plot_width=600, plot_height=250, title="LibDem Voteshare",tools='box_select,lasso_select,pan,wheel_zoom,box_zoom,reset')
p3.scatter('Fieldwork_Midpoint_Date', 'LD', source=source, color='orange')

# Link the x_range of p2 to p1: p2.x_range
p2.x_range = p1.x_range

# Link the y_range of p2 to p1: p2.y_range
p2.y_range = p1.y_range

# Link the x_range of p3 to p1: p3.x_range
p3.x_range = p1.x_range

layout = column(p1,p2,p3)
show(layout)

In [ ]:

links

social